Supervised Learning Achieves Human-Level Performance in MOBA Games: A Case Study of Honor of Kings
This work addresses the challenge of creating human-level AI for complex MOBA games, which is a significant problem for game AI research.
This paper introduces JueWu-SL, a supervised-learning-based AI that achieves human-level performance in MOBA games. The AI integrates macro-strategy and micromanagement into neural networks, performing competitively at the High King player level in 5v5 Honor of Kings games.
We present JueWu-SL, the first supervised-learning-based artificial intelligence (AI) program that achieves human-level performance in playing multiplayer online battle arena (MOBA) games. Unlike prior attempts, we integrate the macro-strategy and the micromanagement of MOBA-game-playing into neural networks in a supervised and end-to-end manner. Tested on Honor of Kings, the most popular MOBA at present, our AI performs competitively at the level of High King players in standard 5v5 games.